Neurol Sci (2015) 36:1889–1895 DOI 10.1007/s10072-015-2276-0
ORIGINAL ARTICLE
How many segments are there in an orange: normative data for the new Cognitive Estimation Task in an Italian population Federica Scarpina1,2 • Guido E. D’Aniello1 • Alessandro Mauro2,3 Gianluca Castelnuovo1,4 • Sarah E. MacPherson5,6
•
Received: 16 December 2014 / Accepted: 3 June 2015 / Published online: 12 June 2015 Ó Springer-Verlag Italia 2015
Abstract The Cognitive Estimation Test (CET) is widely used by clinicians to assess frontal executive dysfunction. In the present work, the Italian standardization of a new version of the CET is provided. This version consists of two 9-item parallel forms (A and B) that were administered to two hundred and twenty-seven healthy Italian male and female participants aged between 19 and 91 years with 5–24 years of full-time education. Performance on the CET was not related to age or level of education; both forms showed a male CET advantage. The new CET is a useful tool for clinicians and researchers to administer the CET more than once without practice effects, which is considered important when assessing frontal executive abilities.
& Federica Scarpina [email protected] 1
Psychology Research Laboratory, IRCCS Istituto Auxologico Italiano, Ospedale San Giuseppe, Via Cadorna 90, 28824 Piancavallo, Oggebbio, VCO, Italy
2
‘‘Rita Levi Montalcini’’ Department of Neuroscience, University of Turin, Turin, Italy
3
Division of Neurology and Neuro-Rehabilitation, IRCCS Istituto Auxologico Italiano, Ospedale San Giuseppe, Piancavallo, VCO, Italy
4
Department of Psychology, Universita` Cattolica del Sacro Cuore, Milan, Italy
5
Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK
6
Department of Psychology, University of Edinburgh, Edinburgh, UK
Keywords Neuropsychological test Cognitive estimation Executive function Frontal lobe
Introduction Cognitive estimation refers to the ability to apply reasoning strategies in order to answer questions that individuals do not usually know the exact answer to. Producing appropriate responses is thought to rely on the ability to select an appropriate cognitive plan, carry out the selected plan, and check any putative answer obtained [1]. Shallice and Evans [1] developed the first Cognitive Estimation Task (CET) in an attempt to assess these cognitive abilities, many of which are executive in nature. Other CET versions have been developed subsequently for administration in other countries [2–5]; nowadays, it is a widely used test of executive function [6] in neurological and psychiatric conditions ([7–22] see [23] for a review). Recently, MacPherson et al. [7] devised an up-to-date version of the CET providing two parallel forms with 9 questions related to landmarks, people, and objects that individuals from all countries should be familiar with. This new version permits repeated assessment of cognitive estimation abilities in both clinical and experimental settings (e.g., before and after a rehabilitation or pharmacological program). Moreover, it explicitly provides participants with the opportunity to change their responses if they feel that the responses are inappropriate and removes the bizarreness index which was reported in the previous Italian CET version [1, 5]. The aim of the present study was to provide normative data for the new version of CET [7] in a large Italian sample, to evaluate the effects of age, education, and
123
1890
Neurol Sci (2015) 36:1889–1895
gender on performance and to calculate inferential cut-off scores.
distribution of the demographic information of the participants according to age, education, and gender.
Methods
Procedure
Participants
Cognitive Estimation Task (CET)
Two hundred and twenty-seven healthy Italian volunteers (106 men, 121 women) aged between 19 and 91 years (M = 47.37 years, SD 17.13 years) were recruited for this study. They were IRCCS Istituto Auxologico Italiano—Ospedale San Giuseppe employees and their relatives and did not receive reimbursement for their participation. Their level of full-time education ranged between 5 and 24 years (M = 14.95 years, SD 4.35 years). Participants were grouped into different age groups: 19–29, 30–39, 40–49, 50–59, 60–69, and 70–91 years and different levels of education: 3–5 years (primary education), 6–8 years (secondary education), 9–13 years (high school education), and more than 13 years (university education). None of the participants had any previous history of neurological or psychiatric disorders or alcohol abuse. All participants were native Italian speakers. Written consent was obtained according to the Declaration of Helsinki. Table 1 provides the
Participants were administered both 9-item parallel forms of the CET [7]. Each question was read aloud by the experimenter and participants answered verbally. Participants were asked to make a sensible guess or estimate in response to each item and were told that most questions did not have a precise answer, but, if they did, it was unlikely that participants would know the answer. Participants could take as much time as they needed to produce their estimates and answer the items using their chosen unit of measurement. Participants were given the opportunity to change their response if they decided that their first response was not a reasonable estimate.
Table 1 Distribution of the participants’ demographic characteristics according to age, education, and gender Age 19–29
30–39
40–49
50–59
60–69
70–91
Data analysis Spearman’s correlation coefficients were calculated to examine the relationship between CET performance and age, gender, and education. Separate linear regression analyses were then carried out to determine whether any of these demographic characteristics were significant predictors of performance on the CET. Successively, the adjusted scores for both versions of the CET were then converted into Equivalent Scores (ES) [24].
Total
Education 3–5 M
Results 0
0
0
0
0
1
1
0
0
0
1
1
3
5
M
0
1
0
4
5
0
10
F
0
1
2
5
4
2
14
F 6–8
9–13 M
6
6
3
8
9
3
35
F
5
2
6
16
5
6
40
M
13
18
6
17
4
2
60
F
23
14
9
10
4
2
62
[13
Total M
19
25
9
29
18
6
106
F
28
17
17
32
14
13
121
Age and education were reported in years M male, F female
123
Cognitive Estimation Task (CET) Firstly, responses were converted into the same unit of measurement. Outliers for each item’s response were then removed using the interquartile range formula. Low values that fell below the lower quartile by 1.5 or more times the interquartile range and high values that fell above the upper quartile by 1.5 or more times the interquartile range were removed. Actual responses were then scored 0, 1, 2, or 3 points based on the percentiles. A score of 0 was awarded for responses that were deemed normal and fell between the 20th and 80th percentiles. One point was awarded for responses considered quite extreme and that were equal to or more than the 10th percentile but less than the 20th percentile or more than the 80th percentile but less than or equal to the 90th percentile. Two points were awarded to
Neurol Sci (2015) 36:1889–1895
1891
responses considered extreme that were more than or equal to the 5th percentile but less than the 10th percentile or more than the 90th percentile but less than or equal to the 95th percentile. Finally, 3 points were awarded to very extreme responses that were less than the 5th or more than the 95th percentile. Where a response corresponded to more than one percentile category, the response was awarded the fewer number of points (e.g., a score of 8 on item 5 of CET A was awarded a score of zero rather than 1). Table 2 demonstrates the percentile ranges for the 227 healthy participants’ actual responses on the individual CET A and B items. Participants could obtain a total score between 0 and 27 for each 9-item CET where the higher the score, the greater the number of responses that deviated from the group. The mean error score for the CET A for the entire sample was 5.66 out of a possible 27 (SD 3.90, median 5.00, range 0–18), and the mean error score for the CET B for the entire sample was 5.36 out of a possible 27 (SD 3.31, median 5.00, range 0–19). The means, standard deviations, and the minimum and maximum values for the actual responses provided for each CET version are shown in Table 3. Both versions A and B of the CET had low reliability and Cronbach’s a = 0.28 and 0.23, respectively. The Guttman split-half reliability coefficient was 0.31 and 0.26, respectively. Spearman’s rank order correlations showed that performance on versions A and B of the CET correlated significantly (r = .30, p \ .0001). Tables 4 and 5 demonstrate the means and standard deviations for the 227 participants performing both versions of the CET according to age group, gender, and level of education. Spearman’s rank order correlations revealed that performance on the CET did not correlate with age (version A: p = .24; version B: p = .80). Performance on the CET did significantly correlate with education on version A of the CET (r = -.14, p \ .05) and approached significance with version B (p = .07). In terms of gender, Mann–Whitney U Tests revealed a
significant main effect of gender on CET A (U = 5060.50, z = -2.75, p \ .01) and CET B (U = 4842.50, z = -3.20, p \ .01) where male participants produced significantly lower CET scores than female participants (version A: M = 4.98, SD 3.17; M = 6.26, SD 3.65, respectively; and version B: M = 4.58, SD 2.88; M = 6.05, SD 3.51, respectively). Linear regression analyses were then conducted for each version of the CET to examine whether age, gender, and years of education significantly contributed to performance on the task. For version A, the analysis revealed a statistically significant model that explains 6.3 % of the variance on the CET with only gender significantly influencing performance (p \ .05). The linear regression equation upon gender is y = 1.17 9 (1 = male or 2 = female) ? 4.36. For version B, a statistically significant model explained 6.0 % of the variance on the CET again with only gender significantly influencing performance (p \ .01). The linear model that explains the variance on the CET is y = 1.40 9 (1 = male or 2 = female) ? 3.92. A correction for gender should be applied to scores on both versions of the CET as the correction is adequate to move an individual’s score from the normal to the impaired range, and vice versa. The correction to achieve adjusted gender scores is calculated as follows: the unstandardized regression coefficient for gender 9 (gender - mean gender). In version A, the adjustment for gender is ?0.62 for males and -0.55 for females. The distribution of the CET error scores for version A of the CET adjusted for gender is as follows: mean 5.66, median 5.45, SD 3.43, minimum = -1, maximum = 17, and interquartile range 4.00. In version B, the adjustment for gender is ?0.74 for males and -0.66 for females. The distribution of the CET error scores for version B of the CET adjusted for gender is as follows: mean 5.36, median 5.34, SD 3.22, minimum = -1, maximum = 18, and interquartile range 5.00. For CET version A, any raw score above the 95th percentile (above 12) should be considered impaired. For CET
Table 2 Percentiles for individual items on versions A and B of the CET Item
CET A percentiles
CET B percentiles
5th
10th
20th
80th
90th
95th
5th
10th
20th
80th
90th
95th
1
100.00
118.00
130.00
200.00
250.00
250.00
2.00
3.00
3.00
6.00
10.00
10.00
2
30.00
35.00
40.00
50.00
60.00
60.00
25.00
34.00
40.00
120.00
130.00
150.00
3
30.00
40.00
50.00
80.00
85.00
100.00
15.75
30.00
50.00
200.00
210.00
230.00
4
5.00
6.00
8.00
20.00
25.00
30.00
30.00
31.60
40.00
84.40
100.00
110.00
5
7.00
8.00
8.00
12.00
12.00
16.00
103.00
104.00
105.00
114.00
115.00
120.00
6
10.00
11.00
15.00
20.00
24.00
25.00
44.50
50.00
60.00
90.00
100.00
100.00
7
50.00
60.00
80.00
120.00
130.00
150.00
30.00
40.00
50.00
100.00
107.00
120.00
8
100.00
120.00
130.00
180.00
200.00
200.00
6.00
10.00
15.00
50.00
60.00
70.30
9
30.00
35.00
40.00
60.00
70.00
80.00
250.00
280.00
300.00
350.00
360.00
380.00
123
1892
Neurol Sci (2015) 36:1889–1895
Table 3 The means, standard deviations, and minimum and maximum values for versions A and B of the CET Category
Unit
Mean
SD
Min
Max
Version A What is the maximum speed of a Harley–Davidson motorbike?
Speed
km/h
169.89
47.25
60
300
What is the length of the average newborn baby?
Length
cm
47.06
8.16
25
65
How fast do race horses run?
Speed
km/h
62.09
20.97
7
120
What is the average jogging speed?
Speed
km/h
13.65
7.18
3
30
How many segments are there in an orange?
Quantity
Segments
10.28
2.39
4
18
What is the length of a new pencil?
Length
cm
17.40
4.26
8
25
What is the maximum speed of a cheetah?
Speed
km/h
100.35
29.60
20
180
What is the length of an average men’s mountain bike?
Length
cm
158.54
30.72
100
240
How many keys are there on a standard computer keyboard?
Quantity
Keys
48.63
13.61
15
90
Version B What is the average walking speed of the typical healthy adult man?
Speed
km/h
How long is the average tie?
Length
cm
5.24
2.24
1
10
83.33
38.59
1
200
What is the fastest tennis serve?
Speed
km/h
117.08
70.80
1
360
How many keys are there on a standard piano?
Quantity
What is the age of the oldest person in your country?
Quantity
Keys
63.24
26.09
11
140
Age
109.69
4.68
100
What is the length of an average man’s spine?
Length
121
cm
71.29
17.89
30
110
What is the maximum speed of a cyclist?
Speed
km/h
71.55
27.96
10
150
How many strings are there on a harp?
Quantity
Strings
31.56
19.88
3
90
What is the maximum speed of a Formula 1 car?
Speed
km/h
321.40
36.07
230
400
Table 4 Means with standard deviations in parentheses per age, gender, and education group for 227 Italian participants performing versions A and B of the CET Education
3–5
Gender
M F
6–8
M F
9–13
M F
[13
M F
CET A
CET B
Age
Age
19–29
30–39
40–49
50–59
60–69
70–91
19–29
30–39
40–49
50–59
60–69
70–91
–
–
–
–
–
3.00
–
–
–
–
–
4.00
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
5.00
4.00
13.33
–
–
–
6.00
8.00
9.67
–
–
–
–
–
(4.04)
–
–
–
–
–
(5.51)
–
1.00
–
4.00
7.40
–
–
5.00
–
4.50
4.40
–
–
–
–
(2.00)
(5.50)
–
–
–
–
(4.12)
(2.70)
– 6.50
–
7.00
4.00
4.40
7.00
9.50
–
9.00
3.50
6.20
8.75
–
–
(1.41)
(2.07)
(5.94)
(4.95)
–
–
(3.54)
(2.17)
(2.22)
(0.71)
6.67
5.83
4.33
3.88
5.78
4.00
5.83
4.50
3.33
5.63
2.89
4.33
(3.20)
(2.99)
(3.21)
(2.30)
(2.64)
(4.00)
(4.31)
(2.17)
(3.06)
(3.29)
(2.32)
(1.53)
7.60
5.50
5.83
5.25
7.80
9.50
9.40
2.50
6.17
5.94
5.20
8.83
(4.62)
(2.12)
(2.48)
(2.21)
(3.96)
(4.68)
(1.95)
(0.71)
(2.99)
(2.64)
(2.49)
(5.42)
6.15 (3.26)
4.17 (2.57)
4.17 (2.40)
4.24 (3.46)
6.50 (3.00)
5.00 (7.07)
5.08 (2.75)
4.78 (3.17)
4.67 (3.88)
4.41 (2.40)
3.50 (3.00)
5.50 (3.54)
5.96
4.86
5.22
5.50
7.50
11.50
5.17
6.00
4.33
5.60
4.00
8.50
(3.27)
(3.72)
(3.19)
(3.21)
(2.38)
(4.95)
(3.08)
(4.61)
(3.24)
(3.50)
(3.74)
(3.54)
Age and education were reported in years
123
Neurol Sci (2015) 36:1889–1895
1893
Table 5 Percentiles of the distribution of the adjusted gender error scores on the parallel versions of the CET Percentiles
CET A
CET B
Percentiles
CET A
CET B
5th
0
1
55th
5
5
10th
2
1
60th
6
6
15th
2
2
65th
6
6
20th
3
2
70th
7
7
25th
3
3
75th
8
7
30th
4
3
80th
8
8
35th 40th
4 4
4 4
85th 90th
9 10
9 9
45th
5
5
95th
12
11
50th
5
5
100th
17
18
The possible scores range from zero (best performance) to 27 (worst performance) Table 6 Equivalent scores for the parallel versions of the CET Equivalent score
CET A
CET B
0
14
13
1
11–13
11–12
2
9–10
8–10
3
6–8
6–7
4
B5
B5
An equivalent score of 0 is considered impaired and an equivalent score of 1 is considered borderline
version B, any gender-adjusted score above the 95th percentile (above 11) should be considered impaired. Table 6 provides the percentiles of the distribution of the CET A and the CET B scores adjusted for gender. The adjusted scores for both versions of the CET were then converted into Equivalent Scores (ES) to allow comparison between CET performance and other clinical tests normed on the Italian population [24, 25]. Given that the CET involves error scores where the higher the score, the poorer the performance, pathological performance was derived by scores higher than the one-sided nonparametric tolerance limit of adjusted scores for 95 % of the population with 95 % confidence [26] and scored 0. In CET version A, this separates 5 participants from the total of 227 (i.e., 2 % of the sample). Adjusted scores below the median value were awarded an ES of 4. The deviation between the 95 % tolerance limit and the median on the normal curve was 2.38 which was divided into 3 sections: 0 to 0.79, 0.79 to 1.58, and 1.58 to 2.38 containing, respectively, 27, 16, and 5 % of the normal distribution. Again, in CET version B, the one-sided nonparametric tolerance limit of adjusted scores for 95 % of the population with 95 % confidence separated 5 participants from the total of 227 (i.e., 2 % of the sample). Adjusted scores below the median value were awarded an ES of 4. The deviation
between the 95 % tolerance limit and the median on the normal curve was 2.29 which was divided into 3 sections: 0 to 0.76, 0.76 to 1.52, and 1.52 to 2.29 containing, respectively, 27, 15, and 6 % of the normal distribution. Table 6 provides the ES scores for the CET A and B.
Discussion The main aim of this study was to provide Italian normative data for the new parallel versions of the CET which allow participants to be assessed on more than one occasion without practice effects [7]. According to the analyses, no correction for age or years of education is necessary for either version of the CET; however, it is necessary to adjust a participant’s error score for gender in versions A and B. Any gender-adjusted score above 12 (for version A) and above 11 (for version B) should be considered impaired. Unlike the British normative data [7], the successful performance of Italian participants on the new CET parallel forms was not associated with increasing age or years of education. While these findings differ from British individuals performing the same CET versions [7], they are in line with the previous Italian version of the CET [5], where performance was associated only with gender, but not with age or education. Other studies in the literature have also reported no influence of age on CET performance [27–29]. It may be that cognitive estimation preservation with age is related to the durability of crystallized intelligence in aging [30]. Recently, Gansler and colleagues [31] administered a revised version of the original CET [1] to 216 healthy participants and found that crystallized intelligence best predicted CET performance, but further work is needed to clarify this issue. The CET advantage found in our Italian male participants has previously been reported in other CET studies [5, 8, 31], including the British sample administered the same CET [7]. While previous CET studies have shown that higher levels of education are associated with lower CET error scores [5, 7, 8], performance on this Italian version was not related to education. Our participants had a wide range of educational levels (ranging from 5 to 18 years and more) compared to previous studies that have tended to recruit individuals that do not have as low a level of education [7], have only high or low education levels [2, 27] or the participants’ education levels are not specified [28] and do report education effects. The current study cannot rule out the possibility that CET performance in the Italian sample is related to general intellectual abilities given that educational level is only a crude measure of intellect. Future work should investigate the role of general intellectual abilities on CET performance in an Italian sample.
123
1894
Neurol Sci (2015) 36:1889–1895
The internal consistency of the items within the CET was very low, as have also been found in other studies [28, 32], suggesting that the items essentially measure different constructs and the complex nature of the CET. There is also an ongoing debate in the literature regarding whether the CET should be considered a measure of global cognitive abilities rather than frontal executive abilities [8]. For the moment, given that clinicians and researchers continue to widely use the CET as a quick and easy assessment of executive dysfunction, the provision of Italian normative data for these new parallel CET versions is timely. Acknowledgments We thank all the participants who took part in the standardization of the new CET. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Shallice T, Evans ME (1978) The involvement of the frontal lobes in cognitive estimation. Cortex 14:294–303. doi:10.1016/ S0010-9452(78)80055-0 2. Axelrod BN, Millis SR (1994) Preliminary Standardization of the Cognitive Estimation Test. Assessment 1:269–274. doi:10.1177/ 107319119400100307 3. Levinoff EJ, Phillips NA, Verret L, Babins L, Kelner N, Akerib V, Chertkow H (2006) Cognitive estimation impairment in Alzheimer disease and mild cognitive impairment. Neuropsychology 20:123–132. doi:10.1037/0894-4105.20.1.123 4. Bullard SE, Fein D, Gleeson MK, Tischer N, Mapou RL, Kaplan E (2004) The Biber Cognitive Estimation Test. Arch Clin Neuropsychol 19:835–846. doi:10.1016/j.acn.2003.12.002 5. Della Sala S, MacPherson SE, Phillips LH, Sacco L, Spinnler H (2003) How many camels are there in Italy? Cognitive estimates standardised on the Italian population. Neurol Sci 24:10–15. doi:10.1007/s100720300015 6. Strauss E, Sherman EMS, Spreen O (2006) A compendium of neuropsychological tests. Oxford University Press, Oxford 7. MacPherson SE, Wagner GP, Murphy P, Bozzali M, Cipolotti L, Shallice T (2014) Bringing the Cognitive Estimation Task into the 21st Century: normative data on two new parallel forms. PLoS One 9(3):e92554. doi:10.1371/journal.pone.0092554 8. Taylor R, O’Carroll R (1995) Cognitive estimation in neurological disorders. Br J Clin Psychol 34:223–228. doi:10.1111/j.20448260.1995.tb01456.x 9. Goldstein FC, Green J, Presley RM, O’Jile J, Freeman A, Watts R, Green RC (1996) Cognitive estimation in patients with Alzheimer’s disease. Neuropsychiatry Neuropsychol Behav Neurol 9(1):35–42 10. Brand M, Fujiwara E, Kalbe E, Steingass HP, Kessler J, Markowitsch HJ (2003) Cognitive estimation and affective judgments in alcoholic Korsakoff patients. J Clin Exp Neuropsychol 25(3):324–334. doi:10.1076/jcen.25.3.324.13802 11. Brand M, Kalbe E, Fujiwara E, Huber M, Markowitsch HJ (2003) Cognitive estimation in patients with probable Alzheimer’s disease and alcoholic Korsakoff patients. Neuropsychologia 41:575–584. doi:10.1016/S0028-3932(02)00183-5
123
12. Della Sala S, MacPherson SE, Phillips LH, Sacco L, Spinnler H (2004) The role of semantic knowledge on the cognitive estimation task. Evidence from Alzheimer’s disease and healthy adult aging. J Neurol 251:156–164. doi:10.1007/s00415-0040292-8 13. Kopelman MD (1991) Frontal dysfunction and memory deficits in the alcoholic Korsakoff syndrome and Alzheimer-Type dementia. Brain 114:117–137. http://dx.doi.org/117-137 14. Mendez MF, Doss CD, Cherrier MM (1998) Use of the Cognitive Estimation Test to discriminate frontotemporal dementia from Alzheimer’s disease. J Geriatr Psychiatry Neurol 11:2–6. doi:10. 1177/089198879801100102 15. Brand M, Fujiwara E, Kalbe E, Steingass HP, Kessler J et al (2003) Cognitive estimation and affective judgments in alcoholic Korsakoff patients. J Clin Exp Neuropsychol 25:324–334. doi:10. 1076/jcen.25.3.324.13802 16. Kopelman MD, Stanhope N, Kingsley D (1999) Retrograde amnesia in patients with diencephalic, temporal lobe or frontal lesions. Neuropsychologia 37:939–958. doi:10.1016/S00283932(98)00143-2 17. Billino J, Brand M, Roesler A (2008) Cognitive estimation in patients with early subcortical vascular dementia. Int J Geriatr Psychiatry 23:982–983. doi:10.1002/gps.1985 18. Leng NR, Parkin AJ (1988) Double dissociation of frontal dysfunction in organic amnesia. Br J Clin Psychol 27:359–362 19. Barabassy A, Beinhoff U, Riepe MW (2010) Cognitive estimation in aged patients with major depressive disorder. Psychiatry Res 176:26–29. doi:10.1016/j.psychres.2008.06.045 20. Freeman MR, Ryan JJ, Lopez SJ, Mittenberg W (1995) Cognitive estimation in traumatic brain injury: relationships with measures of intelligence, memory, and affect. Int J Neurosci 83:269–273. doi:10.3109/00207459508986343 21. Giaglis G, Giannakou M, Fitsioris X, Kosmidis MH, Didaskalou E, Tsiptsios I (2008) Impaired cognitive estimation in non-demented patients with Parkinson’s disease. Eur J Neurol 15:290-290 22. Appollonio IM, Russo A, Isella V, Forapani E, Villa ML, Piolti R, Frattola L (2003) Cognitive [correction of cognitive] estimation: comparison of two tests in nondemented parkinsonian patients. Neurol Sci 24:153–154. doi:10.1007/s10072-0030105-3 23. Wagner GP, MacPherson SE, de Mattos Pimenta Parente MA, Trentini CM (2011) Cognitive estimation abilities in healthy and clinical populations: the use of the Cognitive Estimation Test (CET). Neurol Sci 32(2):203–210. doi:10.1007/s10072-010-0449-4 24. Capitani E, Laiacona M (1988) Aging and psychometric diagnosis of intellectual impairment: some considerations on test scores and their use. Dev Neuropsych 4(4):325–330. doi:10.1080/ 87565648809540416 25. Capitani E, Laiacona M (1997) Composite neuropsychological batteries and demographic correction: standardization based on equivalent scores, with a review of published data. J Clin Exp Neuropsychol 19(6):795–806. doi:10.1080/01688639708403761 26. Wilks SS (1941) Determination of sample size for setting tolerance limits. Ann Math Stat 12(91):96. doi:10.1214/aoms/1177731788 27. O’Carroll R, Egan V, MacKenzie DM (1994) Assessing cognitive estimation. Br J Clin Psychol 33:193–197. doi:10.1111/j.20448260.1994.tb01110.x 28. Gillespie DC, Evans RI, Gardener EA, Bowen A (2002) Performance of older adults on tests of cognitive estimation. J Clin Exp Neuropsychol 24:286–293. doi:10.1076/jcen.24.3.286.988 29. Crawford JR, Bryan J, Luszcz MA, Obonsawin C, Stewart L (2000) The executive decline hypothesis of cognitive aging: do
Neurol Sci (2015) 36:1889–1895 executive deficits qualify as differential deficits and do they mediate age-related memory decline? Aging Neuropsychol Cogn 7:9–31. doi:10.1076/anec.7.1.9.806 30. Cattell RB (1963) Theory of fluid and crystallized intelligence: an initial experiment. J Edu Psychol 105:105–111. doi:10.1037/ h0046743 31. Gansler DA, Varvaris M, Swenson L, Schretlen DJ (2014) Cognitive estimation and its assessment. J Clin Exp Neuropsychol 36(6):559–568. doi:10.1080/13803395.2014.915933
1895 32. Spencer RJ, Johnson-Greene D (2009) The Cognitive Estimation Test (CET): psychometric limitations in neurorehabilitation populations. J Clin Exp Neuropsychol 31:373–377. doi:10.1080/ 13803390802206398
123
ORIGINAL ARTICLE
How many segments are there in an orange: normative data for the new Cognitive Estimation Task in an Italian population Federica Scarpina1,2 • Guido E. D’Aniello1 • Alessandro Mauro2,3 Gianluca Castelnuovo1,4 • Sarah E. MacPherson5,6
•
Received: 16 December 2014 / Accepted: 3 June 2015 / Published online: 12 June 2015 Ó Springer-Verlag Italia 2015
Abstract The Cognitive Estimation Test (CET) is widely used by clinicians to assess frontal executive dysfunction. In the present work, the Italian standardization of a new version of the CET is provided. This version consists of two 9-item parallel forms (A and B) that were administered to two hundred and twenty-seven healthy Italian male and female participants aged between 19 and 91 years with 5–24 years of full-time education. Performance on the CET was not related to age or level of education; both forms showed a male CET advantage. The new CET is a useful tool for clinicians and researchers to administer the CET more than once without practice effects, which is considered important when assessing frontal executive abilities.
& Federica Scarpina [email protected] 1
Psychology Research Laboratory, IRCCS Istituto Auxologico Italiano, Ospedale San Giuseppe, Via Cadorna 90, 28824 Piancavallo, Oggebbio, VCO, Italy
2
‘‘Rita Levi Montalcini’’ Department of Neuroscience, University of Turin, Turin, Italy
3
Division of Neurology and Neuro-Rehabilitation, IRCCS Istituto Auxologico Italiano, Ospedale San Giuseppe, Piancavallo, VCO, Italy
4
Department of Psychology, Universita` Cattolica del Sacro Cuore, Milan, Italy
5
Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK
6
Department of Psychology, University of Edinburgh, Edinburgh, UK
Keywords Neuropsychological test Cognitive estimation Executive function Frontal lobe
Introduction Cognitive estimation refers to the ability to apply reasoning strategies in order to answer questions that individuals do not usually know the exact answer to. Producing appropriate responses is thought to rely on the ability to select an appropriate cognitive plan, carry out the selected plan, and check any putative answer obtained [1]. Shallice and Evans [1] developed the first Cognitive Estimation Task (CET) in an attempt to assess these cognitive abilities, many of which are executive in nature. Other CET versions have been developed subsequently for administration in other countries [2–5]; nowadays, it is a widely used test of executive function [6] in neurological and psychiatric conditions ([7–22] see [23] for a review). Recently, MacPherson et al. [7] devised an up-to-date version of the CET providing two parallel forms with 9 questions related to landmarks, people, and objects that individuals from all countries should be familiar with. This new version permits repeated assessment of cognitive estimation abilities in both clinical and experimental settings (e.g., before and after a rehabilitation or pharmacological program). Moreover, it explicitly provides participants with the opportunity to change their responses if they feel that the responses are inappropriate and removes the bizarreness index which was reported in the previous Italian CET version [1, 5]. The aim of the present study was to provide normative data for the new version of CET [7] in a large Italian sample, to evaluate the effects of age, education, and
123
1890
Neurol Sci (2015) 36:1889–1895
gender on performance and to calculate inferential cut-off scores.
distribution of the demographic information of the participants according to age, education, and gender.
Methods
Procedure
Participants
Cognitive Estimation Task (CET)
Two hundred and twenty-seven healthy Italian volunteers (106 men, 121 women) aged between 19 and 91 years (M = 47.37 years, SD 17.13 years) were recruited for this study. They were IRCCS Istituto Auxologico Italiano—Ospedale San Giuseppe employees and their relatives and did not receive reimbursement for their participation. Their level of full-time education ranged between 5 and 24 years (M = 14.95 years, SD 4.35 years). Participants were grouped into different age groups: 19–29, 30–39, 40–49, 50–59, 60–69, and 70–91 years and different levels of education: 3–5 years (primary education), 6–8 years (secondary education), 9–13 years (high school education), and more than 13 years (university education). None of the participants had any previous history of neurological or psychiatric disorders or alcohol abuse. All participants were native Italian speakers. Written consent was obtained according to the Declaration of Helsinki. Table 1 provides the
Participants were administered both 9-item parallel forms of the CET [7]. Each question was read aloud by the experimenter and participants answered verbally. Participants were asked to make a sensible guess or estimate in response to each item and were told that most questions did not have a precise answer, but, if they did, it was unlikely that participants would know the answer. Participants could take as much time as they needed to produce their estimates and answer the items using their chosen unit of measurement. Participants were given the opportunity to change their response if they decided that their first response was not a reasonable estimate.
Table 1 Distribution of the participants’ demographic characteristics according to age, education, and gender Age 19–29
30–39
40–49
50–59
60–69
70–91
Data analysis Spearman’s correlation coefficients were calculated to examine the relationship between CET performance and age, gender, and education. Separate linear regression analyses were then carried out to determine whether any of these demographic characteristics were significant predictors of performance on the CET. Successively, the adjusted scores for both versions of the CET were then converted into Equivalent Scores (ES) [24].
Total
Education 3–5 M
Results 0
0
0
0
0
1
1
0
0
0
1
1
3
5
M
0
1
0
4
5
0
10
F
0
1
2
5
4
2
14
F 6–8
9–13 M
6
6
3
8
9
3
35
F
5
2
6
16
5
6
40
M
13
18
6
17
4
2
60
F
23
14
9
10
4
2
62
[13
Total M
19
25
9
29
18
6
106
F
28
17
17
32
14
13
121
Age and education were reported in years M male, F female
123
Cognitive Estimation Task (CET) Firstly, responses were converted into the same unit of measurement. Outliers for each item’s response were then removed using the interquartile range formula. Low values that fell below the lower quartile by 1.5 or more times the interquartile range and high values that fell above the upper quartile by 1.5 or more times the interquartile range were removed. Actual responses were then scored 0, 1, 2, or 3 points based on the percentiles. A score of 0 was awarded for responses that were deemed normal and fell between the 20th and 80th percentiles. One point was awarded for responses considered quite extreme and that were equal to or more than the 10th percentile but less than the 20th percentile or more than the 80th percentile but less than or equal to the 90th percentile. Two points were awarded to
Neurol Sci (2015) 36:1889–1895
1891
responses considered extreme that were more than or equal to the 5th percentile but less than the 10th percentile or more than the 90th percentile but less than or equal to the 95th percentile. Finally, 3 points were awarded to very extreme responses that were less than the 5th or more than the 95th percentile. Where a response corresponded to more than one percentile category, the response was awarded the fewer number of points (e.g., a score of 8 on item 5 of CET A was awarded a score of zero rather than 1). Table 2 demonstrates the percentile ranges for the 227 healthy participants’ actual responses on the individual CET A and B items. Participants could obtain a total score between 0 and 27 for each 9-item CET where the higher the score, the greater the number of responses that deviated from the group. The mean error score for the CET A for the entire sample was 5.66 out of a possible 27 (SD 3.90, median 5.00, range 0–18), and the mean error score for the CET B for the entire sample was 5.36 out of a possible 27 (SD 3.31, median 5.00, range 0–19). The means, standard deviations, and the minimum and maximum values for the actual responses provided for each CET version are shown in Table 3. Both versions A and B of the CET had low reliability and Cronbach’s a = 0.28 and 0.23, respectively. The Guttman split-half reliability coefficient was 0.31 and 0.26, respectively. Spearman’s rank order correlations showed that performance on versions A and B of the CET correlated significantly (r = .30, p \ .0001). Tables 4 and 5 demonstrate the means and standard deviations for the 227 participants performing both versions of the CET according to age group, gender, and level of education. Spearman’s rank order correlations revealed that performance on the CET did not correlate with age (version A: p = .24; version B: p = .80). Performance on the CET did significantly correlate with education on version A of the CET (r = -.14, p \ .05) and approached significance with version B (p = .07). In terms of gender, Mann–Whitney U Tests revealed a
significant main effect of gender on CET A (U = 5060.50, z = -2.75, p \ .01) and CET B (U = 4842.50, z = -3.20, p \ .01) where male participants produced significantly lower CET scores than female participants (version A: M = 4.98, SD 3.17; M = 6.26, SD 3.65, respectively; and version B: M = 4.58, SD 2.88; M = 6.05, SD 3.51, respectively). Linear regression analyses were then conducted for each version of the CET to examine whether age, gender, and years of education significantly contributed to performance on the task. For version A, the analysis revealed a statistically significant model that explains 6.3 % of the variance on the CET with only gender significantly influencing performance (p \ .05). The linear regression equation upon gender is y = 1.17 9 (1 = male or 2 = female) ? 4.36. For version B, a statistically significant model explained 6.0 % of the variance on the CET again with only gender significantly influencing performance (p \ .01). The linear model that explains the variance on the CET is y = 1.40 9 (1 = male or 2 = female) ? 3.92. A correction for gender should be applied to scores on both versions of the CET as the correction is adequate to move an individual’s score from the normal to the impaired range, and vice versa. The correction to achieve adjusted gender scores is calculated as follows: the unstandardized regression coefficient for gender 9 (gender - mean gender). In version A, the adjustment for gender is ?0.62 for males and -0.55 for females. The distribution of the CET error scores for version A of the CET adjusted for gender is as follows: mean 5.66, median 5.45, SD 3.43, minimum = -1, maximum = 17, and interquartile range 4.00. In version B, the adjustment for gender is ?0.74 for males and -0.66 for females. The distribution of the CET error scores for version B of the CET adjusted for gender is as follows: mean 5.36, median 5.34, SD 3.22, minimum = -1, maximum = 18, and interquartile range 5.00. For CET version A, any raw score above the 95th percentile (above 12) should be considered impaired. For CET
Table 2 Percentiles for individual items on versions A and B of the CET Item
CET A percentiles
CET B percentiles
5th
10th
20th
80th
90th
95th
5th
10th
20th
80th
90th
95th
1
100.00
118.00
130.00
200.00
250.00
250.00
2.00
3.00
3.00
6.00
10.00
10.00
2
30.00
35.00
40.00
50.00
60.00
60.00
25.00
34.00
40.00
120.00
130.00
150.00
3
30.00
40.00
50.00
80.00
85.00
100.00
15.75
30.00
50.00
200.00
210.00
230.00
4
5.00
6.00
8.00
20.00
25.00
30.00
30.00
31.60
40.00
84.40
100.00
110.00
5
7.00
8.00
8.00
12.00
12.00
16.00
103.00
104.00
105.00
114.00
115.00
120.00
6
10.00
11.00
15.00
20.00
24.00
25.00
44.50
50.00
60.00
90.00
100.00
100.00
7
50.00
60.00
80.00
120.00
130.00
150.00
30.00
40.00
50.00
100.00
107.00
120.00
8
100.00
120.00
130.00
180.00
200.00
200.00
6.00
10.00
15.00
50.00
60.00
70.30
9
30.00
35.00
40.00
60.00
70.00
80.00
250.00
280.00
300.00
350.00
360.00
380.00
123
1892
Neurol Sci (2015) 36:1889–1895
Table 3 The means, standard deviations, and minimum and maximum values for versions A and B of the CET Category
Unit
Mean
SD
Min
Max
Version A What is the maximum speed of a Harley–Davidson motorbike?
Speed
km/h
169.89
47.25
60
300
What is the length of the average newborn baby?
Length
cm
47.06
8.16
25
65
How fast do race horses run?
Speed
km/h
62.09
20.97
7
120
What is the average jogging speed?
Speed
km/h
13.65
7.18
3
30
How many segments are there in an orange?
Quantity
Segments
10.28
2.39
4
18
What is the length of a new pencil?
Length
cm
17.40
4.26
8
25
What is the maximum speed of a cheetah?
Speed
km/h
100.35
29.60
20
180
What is the length of an average men’s mountain bike?
Length
cm
158.54
30.72
100
240
How many keys are there on a standard computer keyboard?
Quantity
Keys
48.63
13.61
15
90
Version B What is the average walking speed of the typical healthy adult man?
Speed
km/h
How long is the average tie?
Length
cm
5.24
2.24
1
10
83.33
38.59
1
200
What is the fastest tennis serve?
Speed
km/h
117.08
70.80
1
360
How many keys are there on a standard piano?
Quantity
What is the age of the oldest person in your country?
Quantity
Keys
63.24
26.09
11
140
Age
109.69
4.68
100
What is the length of an average man’s spine?
Length
121
cm
71.29
17.89
30
110
What is the maximum speed of a cyclist?
Speed
km/h
71.55
27.96
10
150
How many strings are there on a harp?
Quantity
Strings
31.56
19.88
3
90
What is the maximum speed of a Formula 1 car?
Speed
km/h
321.40
36.07
230
400
Table 4 Means with standard deviations in parentheses per age, gender, and education group for 227 Italian participants performing versions A and B of the CET Education
3–5
Gender
M F
6–8
M F
9–13
M F
[13
M F
CET A
CET B
Age
Age
19–29
30–39
40–49
50–59
60–69
70–91
19–29
30–39
40–49
50–59
60–69
70–91
–
–
–
–
–
3.00
–
–
–
–
–
4.00
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
5.00
4.00
13.33
–
–
–
6.00
8.00
9.67
–
–
–
–
–
(4.04)
–
–
–
–
–
(5.51)
–
1.00
–
4.00
7.40
–
–
5.00
–
4.50
4.40
–
–
–
–
(2.00)
(5.50)
–
–
–
–
(4.12)
(2.70)
– 6.50
–
7.00
4.00
4.40
7.00
9.50
–
9.00
3.50
6.20
8.75
–
–
(1.41)
(2.07)
(5.94)
(4.95)
–
–
(3.54)
(2.17)
(2.22)
(0.71)
6.67
5.83
4.33
3.88
5.78
4.00
5.83
4.50
3.33
5.63
2.89
4.33
(3.20)
(2.99)
(3.21)
(2.30)
(2.64)
(4.00)
(4.31)
(2.17)
(3.06)
(3.29)
(2.32)
(1.53)
7.60
5.50
5.83
5.25
7.80
9.50
9.40
2.50
6.17
5.94
5.20
8.83
(4.62)
(2.12)
(2.48)
(2.21)
(3.96)
(4.68)
(1.95)
(0.71)
(2.99)
(2.64)
(2.49)
(5.42)
6.15 (3.26)
4.17 (2.57)
4.17 (2.40)
4.24 (3.46)
6.50 (3.00)
5.00 (7.07)
5.08 (2.75)
4.78 (3.17)
4.67 (3.88)
4.41 (2.40)
3.50 (3.00)
5.50 (3.54)
5.96
4.86
5.22
5.50
7.50
11.50
5.17
6.00
4.33
5.60
4.00
8.50
(3.27)
(3.72)
(3.19)
(3.21)
(2.38)
(4.95)
(3.08)
(4.61)
(3.24)
(3.50)
(3.74)
(3.54)
Age and education were reported in years
123
Neurol Sci (2015) 36:1889–1895
1893
Table 5 Percentiles of the distribution of the adjusted gender error scores on the parallel versions of the CET Percentiles
CET A
CET B
Percentiles
CET A
CET B
5th
0
1
55th
5
5
10th
2
1
60th
6
6
15th
2
2
65th
6
6
20th
3
2
70th
7
7
25th
3
3
75th
8
7
30th
4
3
80th
8
8
35th 40th
4 4
4 4
85th 90th
9 10
9 9
45th
5
5
95th
12
11
50th
5
5
100th
17
18
The possible scores range from zero (best performance) to 27 (worst performance) Table 6 Equivalent scores for the parallel versions of the CET Equivalent score
CET A
CET B
0
14
13
1
11–13
11–12
2
9–10
8–10
3
6–8
6–7
4
B5
B5
An equivalent score of 0 is considered impaired and an equivalent score of 1 is considered borderline
version B, any gender-adjusted score above the 95th percentile (above 11) should be considered impaired. Table 6 provides the percentiles of the distribution of the CET A and the CET B scores adjusted for gender. The adjusted scores for both versions of the CET were then converted into Equivalent Scores (ES) to allow comparison between CET performance and other clinical tests normed on the Italian population [24, 25]. Given that the CET involves error scores where the higher the score, the poorer the performance, pathological performance was derived by scores higher than the one-sided nonparametric tolerance limit of adjusted scores for 95 % of the population with 95 % confidence [26] and scored 0. In CET version A, this separates 5 participants from the total of 227 (i.e., 2 % of the sample). Adjusted scores below the median value were awarded an ES of 4. The deviation between the 95 % tolerance limit and the median on the normal curve was 2.38 which was divided into 3 sections: 0 to 0.79, 0.79 to 1.58, and 1.58 to 2.38 containing, respectively, 27, 16, and 5 % of the normal distribution. Again, in CET version B, the one-sided nonparametric tolerance limit of adjusted scores for 95 % of the population with 95 % confidence separated 5 participants from the total of 227 (i.e., 2 % of the sample). Adjusted scores below the median value were awarded an ES of 4. The deviation
between the 95 % tolerance limit and the median on the normal curve was 2.29 which was divided into 3 sections: 0 to 0.76, 0.76 to 1.52, and 1.52 to 2.29 containing, respectively, 27, 15, and 6 % of the normal distribution. Table 6 provides the ES scores for the CET A and B.
Discussion The main aim of this study was to provide Italian normative data for the new parallel versions of the CET which allow participants to be assessed on more than one occasion without practice effects [7]. According to the analyses, no correction for age or years of education is necessary for either version of the CET; however, it is necessary to adjust a participant’s error score for gender in versions A and B. Any gender-adjusted score above 12 (for version A) and above 11 (for version B) should be considered impaired. Unlike the British normative data [7], the successful performance of Italian participants on the new CET parallel forms was not associated with increasing age or years of education. While these findings differ from British individuals performing the same CET versions [7], they are in line with the previous Italian version of the CET [5], where performance was associated only with gender, but not with age or education. Other studies in the literature have also reported no influence of age on CET performance [27–29]. It may be that cognitive estimation preservation with age is related to the durability of crystallized intelligence in aging [30]. Recently, Gansler and colleagues [31] administered a revised version of the original CET [1] to 216 healthy participants and found that crystallized intelligence best predicted CET performance, but further work is needed to clarify this issue. The CET advantage found in our Italian male participants has previously been reported in other CET studies [5, 8, 31], including the British sample administered the same CET [7]. While previous CET studies have shown that higher levels of education are associated with lower CET error scores [5, 7, 8], performance on this Italian version was not related to education. Our participants had a wide range of educational levels (ranging from 5 to 18 years and more) compared to previous studies that have tended to recruit individuals that do not have as low a level of education [7], have only high or low education levels [2, 27] or the participants’ education levels are not specified [28] and do report education effects. The current study cannot rule out the possibility that CET performance in the Italian sample is related to general intellectual abilities given that educational level is only a crude measure of intellect. Future work should investigate the role of general intellectual abilities on CET performance in an Italian sample.
123
1894
Neurol Sci (2015) 36:1889–1895
The internal consistency of the items within the CET was very low, as have also been found in other studies [28, 32], suggesting that the items essentially measure different constructs and the complex nature of the CET. There is also an ongoing debate in the literature regarding whether the CET should be considered a measure of global cognitive abilities rather than frontal executive abilities [8]. For the moment, given that clinicians and researchers continue to widely use the CET as a quick and easy assessment of executive dysfunction, the provision of Italian normative data for these new parallel CET versions is timely. Acknowledgments We thank all the participants who took part in the standardization of the new CET. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Shallice T, Evans ME (1978) The involvement of the frontal lobes in cognitive estimation. Cortex 14:294–303. doi:10.1016/ S0010-9452(78)80055-0 2. Axelrod BN, Millis SR (1994) Preliminary Standardization of the Cognitive Estimation Test. Assessment 1:269–274. doi:10.1177/ 107319119400100307 3. Levinoff EJ, Phillips NA, Verret L, Babins L, Kelner N, Akerib V, Chertkow H (2006) Cognitive estimation impairment in Alzheimer disease and mild cognitive impairment. Neuropsychology 20:123–132. doi:10.1037/0894-4105.20.1.123 4. Bullard SE, Fein D, Gleeson MK, Tischer N, Mapou RL, Kaplan E (2004) The Biber Cognitive Estimation Test. Arch Clin Neuropsychol 19:835–846. doi:10.1016/j.acn.2003.12.002 5. Della Sala S, MacPherson SE, Phillips LH, Sacco L, Spinnler H (2003) How many camels are there in Italy? Cognitive estimates standardised on the Italian population. Neurol Sci 24:10–15. doi:10.1007/s100720300015 6. Strauss E, Sherman EMS, Spreen O (2006) A compendium of neuropsychological tests. Oxford University Press, Oxford 7. MacPherson SE, Wagner GP, Murphy P, Bozzali M, Cipolotti L, Shallice T (2014) Bringing the Cognitive Estimation Task into the 21st Century: normative data on two new parallel forms. PLoS One 9(3):e92554. doi:10.1371/journal.pone.0092554 8. Taylor R, O’Carroll R (1995) Cognitive estimation in neurological disorders. Br J Clin Psychol 34:223–228. doi:10.1111/j.20448260.1995.tb01456.x 9. Goldstein FC, Green J, Presley RM, O’Jile J, Freeman A, Watts R, Green RC (1996) Cognitive estimation in patients with Alzheimer’s disease. Neuropsychiatry Neuropsychol Behav Neurol 9(1):35–42 10. Brand M, Fujiwara E, Kalbe E, Steingass HP, Kessler J, Markowitsch HJ (2003) Cognitive estimation and affective judgments in alcoholic Korsakoff patients. J Clin Exp Neuropsychol 25(3):324–334. doi:10.1076/jcen.25.3.324.13802 11. Brand M, Kalbe E, Fujiwara E, Huber M, Markowitsch HJ (2003) Cognitive estimation in patients with probable Alzheimer’s disease and alcoholic Korsakoff patients. Neuropsychologia 41:575–584. doi:10.1016/S0028-3932(02)00183-5
123
12. Della Sala S, MacPherson SE, Phillips LH, Sacco L, Spinnler H (2004) The role of semantic knowledge on the cognitive estimation task. Evidence from Alzheimer’s disease and healthy adult aging. J Neurol 251:156–164. doi:10.1007/s00415-0040292-8 13. Kopelman MD (1991) Frontal dysfunction and memory deficits in the alcoholic Korsakoff syndrome and Alzheimer-Type dementia. Brain 114:117–137. http://dx.doi.org/117-137 14. Mendez MF, Doss CD, Cherrier MM (1998) Use of the Cognitive Estimation Test to discriminate frontotemporal dementia from Alzheimer’s disease. J Geriatr Psychiatry Neurol 11:2–6. doi:10. 1177/089198879801100102 15. Brand M, Fujiwara E, Kalbe E, Steingass HP, Kessler J et al (2003) Cognitive estimation and affective judgments in alcoholic Korsakoff patients. J Clin Exp Neuropsychol 25:324–334. doi:10. 1076/jcen.25.3.324.13802 16. Kopelman MD, Stanhope N, Kingsley D (1999) Retrograde amnesia in patients with diencephalic, temporal lobe or frontal lesions. Neuropsychologia 37:939–958. doi:10.1016/S00283932(98)00143-2 17. Billino J, Brand M, Roesler A (2008) Cognitive estimation in patients with early subcortical vascular dementia. Int J Geriatr Psychiatry 23:982–983. doi:10.1002/gps.1985 18. Leng NR, Parkin AJ (1988) Double dissociation of frontal dysfunction in organic amnesia. Br J Clin Psychol 27:359–362 19. Barabassy A, Beinhoff U, Riepe MW (2010) Cognitive estimation in aged patients with major depressive disorder. Psychiatry Res 176:26–29. doi:10.1016/j.psychres.2008.06.045 20. Freeman MR, Ryan JJ, Lopez SJ, Mittenberg W (1995) Cognitive estimation in traumatic brain injury: relationships with measures of intelligence, memory, and affect. Int J Neurosci 83:269–273. doi:10.3109/00207459508986343 21. Giaglis G, Giannakou M, Fitsioris X, Kosmidis MH, Didaskalou E, Tsiptsios I (2008) Impaired cognitive estimation in non-demented patients with Parkinson’s disease. Eur J Neurol 15:290-290 22. Appollonio IM, Russo A, Isella V, Forapani E, Villa ML, Piolti R, Frattola L (2003) Cognitive [correction of cognitive] estimation: comparison of two tests in nondemented parkinsonian patients. Neurol Sci 24:153–154. doi:10.1007/s10072-0030105-3 23. Wagner GP, MacPherson SE, de Mattos Pimenta Parente MA, Trentini CM (2011) Cognitive estimation abilities in healthy and clinical populations: the use of the Cognitive Estimation Test (CET). Neurol Sci 32(2):203–210. doi:10.1007/s10072-010-0449-4 24. Capitani E, Laiacona M (1988) Aging and psychometric diagnosis of intellectual impairment: some considerations on test scores and their use. Dev Neuropsych 4(4):325–330. doi:10.1080/ 87565648809540416 25. Capitani E, Laiacona M (1997) Composite neuropsychological batteries and demographic correction: standardization based on equivalent scores, with a review of published data. J Clin Exp Neuropsychol 19(6):795–806. doi:10.1080/01688639708403761 26. Wilks SS (1941) Determination of sample size for setting tolerance limits. Ann Math Stat 12(91):96. doi:10.1214/aoms/1177731788 27. O’Carroll R, Egan V, MacKenzie DM (1994) Assessing cognitive estimation. Br J Clin Psychol 33:193–197. doi:10.1111/j.20448260.1994.tb01110.x 28. Gillespie DC, Evans RI, Gardener EA, Bowen A (2002) Performance of older adults on tests of cognitive estimation. J Clin Exp Neuropsychol 24:286–293. doi:10.1076/jcen.24.3.286.988 29. Crawford JR, Bryan J, Luszcz MA, Obonsawin C, Stewart L (2000) The executive decline hypothesis of cognitive aging: do
Neurol Sci (2015) 36:1889–1895 executive deficits qualify as differential deficits and do they mediate age-related memory decline? Aging Neuropsychol Cogn 7:9–31. doi:10.1076/anec.7.1.9.806 30. Cattell RB (1963) Theory of fluid and crystallized intelligence: an initial experiment. J Edu Psychol 105:105–111. doi:10.1037/ h0046743 31. Gansler DA, Varvaris M, Swenson L, Schretlen DJ (2014) Cognitive estimation and its assessment. J Clin Exp Neuropsychol 36(6):559–568. doi:10.1080/13803395.2014.915933
1895 32. Spencer RJ, Johnson-Greene D (2009) The Cognitive Estimation Test (CET): psychometric limitations in neurorehabilitation populations. J Clin Exp Neuropsychol 31:373–377. doi:10.1080/ 13803390802206398
123
